Mobile Edition
Print
Get the Mag
Weekly UpdatesThe literature on television effects frequently finds that, although television viewers report high levels of exposure to television, memory for what they have seen is very low (Gunter, 1987). Recent research is beginning to explain these findings by demonstrating that the relationships between exposure and attention, and between attention and memory are not simple, direct, or linear.
Many researchers and research traditions have built a convincing body of evidence that exposure does not determine attention (Biocca, 1988; Chaffee & Schleuder, 1986). Rather, exposure seems to be a necessary but not sufficient case for attention to occur. Viewers' needs, intentions, and goals play a large role in determining whether the viewer will pay more or less attention to a message (Gantz, 1978; Geiger & Newhagen, 1993; Gunter, 1987; Levy & Windahl, 1984; Petty, Cacioppo, & Schumann, 1983).
Research also shows that attention levels do not remain constant during viewing of a message; attention frequently varies both between and within programs, individuals, and situations. In particular, attention levels during viewing of a single message have been shown to fluctuate predictably as a function of a television message's structure and content (Lang, 1995; Reeves, Thorson, & Schleuder, 1985; Reeves & Thorson, 1986; Reeves, et al., 1985). This research has demonstrated fairly convincingly that exposure to a message is not a guarantee of attention. Even among "attentive viewers", attention level varies over the course of a viewing session.
Recent research suggests that a similar situation exists for the relationship between attention and memory. Early research often inferred attention by measuring memory - making the assumption that if viewers remembered something then they must have paid attention to it, and if they didn't remember something, it was because they hadn't paid attention to it (Grimes & Meadowcroft, 1995). However, it now appears that many types of television messages elicit quite high levels of "attention" and quite low levels of memory for the content of the message (Gunter, 1987; Thorson, Reeves, & Schleuder, 1985, 1986).
Using the limited capacity approach to television viewing to analyze the relationship between TV's form and content and viewers' attention to and memory for television messages, Lang and her colleagues have shown that many aspects of television can create states of high attention which result in poor memory for television messages (Lang, Bolls, Potter, & Kawahara, 1999; Lang, Newhagen, & Reeves, 1996).
A Limited Capacity Approach to Television Viewing
The limited capacity approach to television viewing (Lang, 1995; Lang & Basil, 1998; Lang, Bolls, Potter, & Kawahara, 1999; Lang, Newhagen, & Reeves, 1996) suggests that viewers' information processing resources are limited. In order to process television messages, television viewers must encode the information contained in the message, retrieve already stored information from long term memory in order to make sense of the incoming message, and store the new information in long term memory. This approach argues that three sub-processes of information processing-encoding, storage, and retrieval-occur continuously, simultaneously, and to some extent automatically while viewers watch television. The viewers' fixed capacity for limited processing resources are flexibly distributed across these three simultaneously occurring processes(1). The distribution of resources is determined both by automatic processes (triggered by content and structural features of the message) and by controlled processes (driven by viewer interests, needs, goals, and motivations).
The task of watching television is performed adequately when all three sub-processes have sufficient resources available to perform at the level desired/required by the viewer. If there are not sufficient resources available, the overall task of watching television will be performed less effectively. When this occurs, it is often the case that one of the sub-processes will be more affected than the others.
During television viewing, viewers are not in control of the pace at which information is presented: the viewer must "keep up" with the message. As a consequence, resources to encode and make sense of a message are allocated automatically in response to the structural and content characteristics of the message. When a message is difficult--either in terms of content or in terms of structure-an increase in the number of resources allocated to encoding and on-line retrieval (the sub-processes primarily involved in sense-making) results. The viewer in this case has fewer resources available for the sub-process of message storage. When this happens, the viewer remembers less of the message. This means that a message may receive very high levels of attention (that is the viewer is working very hard to encode and make sense of the message), but still will be remembered poorly, because the viewer was unable to allocate sufficient processing resources to store the message.
Previous research testing this theory has demonstrated that one can increase viewer attention to the message while decreasing their memory for the message by increasing either the structural complexity of a message (Lang et al., 1999; Yoon, Bolls, & Lang, 1998; Yoon, Bolls, Lang, & Potter, 1997), the content difficulty of a message (Lang, Geiger, Strickwerda, & Sumner, 1993; Thorson & Lang, 1992), or both (Lang et al., 1999).
One way to increase the resources allocated to encoding a television message is by introducing structural features (such as cuts, edits, graphics, and sound effects) into television messages. Many structural features of television elicit what is called an orienting response (Lang, 1990).
The "orienting response" is an involuntary physiological and behavioral response that directs our attention toward new or relevant information in the environment. The orienting response is made up of a set of physiological and behavioral responses which include: turning sensory receptors (eyes, ears, nose) toward the stimulus, lowered heart rate, decreased blood flow to the muscles, alpha wave suppression in the electroencephalogram (EEG), increased skin temperature, increased electrical conductivity of the skin, and increased blood flow to the brain (Lynn, 1966; Lang, Simons, & Balaban, 1997). Orienting responses to television messages have been measured using eyes on screen (Anderson, 1983), EEG (Reeves, et al., 1985), and heart rate (Lang, 1990; Lang, Geiger, Strickwerda, & Sumner, 1993; Lang, Newhagen, & Reeves, 1996; Lang & Thorson, 1992).
The limited capacity approach to television viewing suggests that when a television message elicits an orienting response, this results in an increase in the allocation of processing resources to encoding the information in the message. Several studies (Lang, 1990; Lang, 1991; Reeves, et al., 1985; & Thorson & Lang, 1989) have demonstrated that the formal or structural features of television (such as cuts, movement, and sound changes) both evoke the orienting response and increase the resources allocated to processing messages. Both Lang et al. (1993) and Geiger and Reeves, (1993) demonstrated that secondary task reaction times (often used as a measure of resources allocated to processing) are slower immediately following both cuts and edits in television messages.
At the same time, this research shows an increase in attention and resource allocation elicited by the cut or edit does not lead automatically to an increase in memory for the message. Rather, the short-term or local effect of the structural feature on memory varies as a function of the overall, global cognitive load imposed by the content of the message. The global cognitive load imposed by the message can be varied by manipulating either the structure or the content of the message.
Thorson & Lang (1992) manipulated global cognitive load by varying the difficulty of the material contained in videotaped lectures, then examining the effects of local orienting responses elicited by videographics on memory. They demonstrated that memory for the content of the lectures increased following orienting responses when the global content of the lecture was easy for the viewer (low cognitive load). However, when the global lecture content was difficult, memory for information following the orienting response decreased (high cognitive load).
Similarly, structural features can be used to manipulate both the global and the local cognitive load. Global cognitive load can be manipulated by increasing the number and rate of structural features contained in a message (Lang et al., 1999; Reeves, Thorson, & Schleuder, 1985).
Local cognitive load can be affected by the type of structural feature. Lang and Basil (1998) suggest that one can conceptualize structural features in terms of the amount of new information introduced by the structural feature. Some types of structural features typically occur in conjunction with the introduction of new information. That information may be expected or unexpected, and it may be semantically related or semantically unrelated to the information which preceded the structural feature. Lang and Basil (1998) suggest that when structural features introduce a lot of new information or unexpected or unrelated information, they require more resources to be fully processed. On the other hand, structural features which introduce little new information or expected information will require fewer resources to be fully processed. This suggestion was made based on a review of the research investigating resource allocation, attention, and memory elicited by various types of camera and scene changes in television messages.
Lang et al. (1993) varied the cognitive load of structural features by varying the "relatedness" of the message on either side of a cut. Related cuts were defined as cuts where the information following the cut was narratively and semantically related to the information preceding the cut. Unrelated cuts were defined as cuts where the information following the cut was narratively and semantically unrelated to the information preceding the cut. They hypothesized that unrelated cuts impose a greater cognitive load on the viewer and require more resources to be fully processed. Results indicated that memory for information following the cut was better for related cuts than for unrelated cuts.
FEED